Achievable logoAchievable logo
AP Statistics
Sign in
Sign up
Purchase
Textbook
Practice exams
Support
How it works
Resources
Exam catalog
Mountain with a flag at the peak
Textbook
Introduction
1. One variable data
2. Two variable data
3. Data collection
4. Probability and random variables
4.1 Law of large numbers
4.2 Introduction to probability
4.3 Expected value, variance, and standard deviation
4.4 Binomial distribution
4.5 Geometric distribution
4.6 Cumulative probability distribution
5. Sampling distributions
6. Categorical data
7. Quantitative data
8. Chi-square
9. Linear regression
Wrapping up
Achievable logoAchievable logo
4.5 Geometric distribution
Achievable AP Statistics
4. Probability and random variables
Our AP Statistics course is currently in development and is a work-in-progress.

Geometric distribution

4 min read
Font
Discuss
Share
Feedback

Expected value and variation

Understanding how probability connects to real-world averages starts with the expected value. The definitions below describe three related measures you’ll use throughout this section.

Definitions
Expected value
The theoretical long-run average outcome over a very large number of trials. For discrete random variables:

E(x)=∑x⋅p(x)

where x refers to the value of the outcome and p(x) refers to the probability of the outcome.

Variance
A measure of how spread out values are around the mean. A smaller variance means values tend to be closer to the mean, and a larger variance means values tend to be farther from the mean.

Var(X)=E(X2)−[E(X)]2

Standard deviation
Another measure of variation — the square root of the variance.

Standard deviation=Var(X)​

Sidenote
Note about expected value:

The expected value does not mean the outcome you’re most likely to see in a single trial. In fact, the expected value is often not a value that can actually occur. Instead, it represents the average result you’d get if you repeated the situation a very large number of times — a theoretical weighted average computed using the formula above.

Practice problem

Example:

A student randomly draws a bill out of a hat and will get to keep whichever bill he draws. The hat contains fifty $1 bills, ten $5 bills, and five $20 bills. Find the expected value, variance, and standard deviation.

Starting with expected value:

E(x)​=6550​⋅1+6510​⋅5+655​⋅20=3.076923​

To find variance, first calculate E(X2):

E(X2)​=6550​⋅12+6510​⋅52+655​⋅202=6550​+65250​+652000​=652300​=35.384615​

Var(X)​=E(X2)−[E(X)]2=35.384615−(3.076923)2=25.917159​

Standard deviation​=Var(X)​=25.917159​​

Solution:

(spoiler)

The expected value is $3.08, the variance is approximately 25.92, and the standard deviation is approximately $5.09.

Properties of expected value and standard deviation

For any two random variables X and Y, the following properties hold. These rules make it possible to work with combined random variables without calculating everything from scratch.

E(X+Y)E(X−Y)​=E(X)+E(Y)=E(X)−E(Y)​

If X and Y are independent:

σX±Y​=σX2​+σY2​​

Practice problem

Example:

The customer service department of a service provider receives complaints about the company’s two products: phone and internet. Here are the possible numbers of complaints of each type that the company may receive per day and their probabilities:

# of phone complaints Probability
0 0.3
1 0.5
2 0.2
# of internet complaints Probability
0 0.4
1 0.4
2 0.2

Part a.

Find the expected total number of complaints per day.

Let X represent phone complaints and Y represent internet complaints.

E(X)​=0⋅0.3+1⋅0.5+2⋅0.2=0.9​

E(Y)​=0⋅0.4+1⋅0.4+2⋅0.2=0.8​

E(X+Y)​=E(X)+E(Y)=0.9+0.8=1.7​

Solution:

(spoiler)

The expected number of complaints is 1.7 per day.

Part b.

Assuming independence, what is the standard deviation of the total complaints?

Find the variance of each variable separately, then combine.

E(X2)​=02⋅0.3+12⋅0.5+22⋅0.2=1.3​

Var(X)​=1.3−(0.9)2=0.49​

E(Y2)​=02⋅0.4+12⋅0.4+22⋅0.2=1.2​

Var(Y)​=1.2−(0.8)2=0.56​

Var(X+Y)​=0.49+0.56=1.05​

Standard deviation​=1.05​​

Solution:

(spoiler)

The standard deviation of total complaints is approximately 1.02.

Geometric distribution basics

  • Models number of trials until first success
  • Probability of success per trial: p
  • Probability formula: P(X=k)=(1−p)k−1⋅p

Formulas

  • Expected value: E(X)=p1​
  • Variance: Var(X)=p21−p​
  • Standard deviation: σ=p21−p​​

BITS conditions

  • Binary outcomes: success or failure
  • Independent trials
  • Trials not fixed; count until first success
  • Same probability p for each trial

Technology functions

  • geometpdf(p,x): probability first success on xth trial
  • geometcdf(p,x): probability first success on or before xth trial

Example applications

  • Define X as number of trials until first success (e.g., first third party voter)
  • Use E(X) for expected number of trials
  • Use σ for typical deviation from mean
  • Use P(X=k) or geometpdf for exact probabilities
  • Use geometcdf for cumulative probabilities
  • Use complement rule for probabilities of “at least” scenarios

Summary

  • Key formulas: P(X=k), E(X), Var(X), σ
  • Use technology for quick probability calculations
  • Always check BITS before applying geometric distribution

Sign up for free to take 8 quiz questions on this topic

Previous
Next  | 4.6 Cumulative probability distribution
All rights reserved ©2016 - 2026 Achievable, Inc.

Geometric distribution

Expected value and variation

Understanding how probability connects to real-world averages starts with the expected value. The definitions below describe three related measures you’ll use throughout this section.

Definitions
Expected value
The theoretical long-run average outcome over a very large number of trials. For discrete random variables:

E(x)=∑x⋅p(x)

where x refers to the value of the outcome and p(x) refers to the probability of the outcome.

Variance
A measure of how spread out values are around the mean. A smaller variance means values tend to be closer to the mean, and a larger variance means values tend to be farther from the mean.

Var(X)=E(X2)−[E(X)]2

Standard deviation
Another measure of variation — the square root of the variance.

Standard deviation=Var(X)​

Sidenote
Note about expected value:

The expected value does not mean the outcome you’re most likely to see in a single trial. In fact, the expected value is often not a value that can actually occur. Instead, it represents the average result you’d get if you repeated the situation a very large number of times — a theoretical weighted average computed using the formula above.

Practice problem

Example:

A student randomly draws a bill out of a hat and will get to keep whichever bill he draws. The hat contains fifty $1 bills, ten $5 bills, and five $20 bills. Find the expected value, variance, and standard deviation.

Starting with expected value:

E(x)​=6550​⋅1+6510​⋅5+655​⋅20=3.076923​

To find variance, first calculate E(X2):

E(X2)​=6550​⋅12+6510​⋅52+655​⋅202=6550​+65250​+652000​=652300​=35.384615​

Var(X)​=E(X2)−[E(X)]2=35.384615−(3.076923)2=25.917159​

Standard deviation​=Var(X)​=25.917159​​

Solution:

(spoiler)

The expected value is $3.08, the variance is approximately 25.92, and the standard deviation is approximately $5.09.

Properties of expected value and standard deviation

For any two random variables X and Y, the following properties hold. These rules make it possible to work with combined random variables without calculating everything from scratch.

E(X+Y)E(X−Y)​=E(X)+E(Y)=E(X)−E(Y)​

If X and Y are independent:

σX±Y​=σX2​+σY2​​

Practice problem

Example:

The customer service department of a service provider receives complaints about the company’s two products: phone and internet. Here are the possible numbers of complaints of each type that the company may receive per day and their probabilities:

# of phone complaints Probability
0 0.3
1 0.5
2 0.2
# of internet complaints Probability
0 0.4
1 0.4
2 0.2

Part a.

Find the expected total number of complaints per day.

Let X represent phone complaints and Y represent internet complaints.

E(X)​=0⋅0.3+1⋅0.5+2⋅0.2=0.9​

E(Y)​=0⋅0.4+1⋅0.4+2⋅0.2=0.8​

E(X+Y)​=E(X)+E(Y)=0.9+0.8=1.7​

Solution:

(spoiler)

The expected number of complaints is 1.7 per day.

Part b.

Assuming independence, what is the standard deviation of the total complaints?

Find the variance of each variable separately, then combine.

E(X2)​=02⋅0.3+12⋅0.5+22⋅0.2=1.3​

Var(X)​=1.3−(0.9)2=0.49​

E(Y2)​=02⋅0.4+12⋅0.4+22⋅0.2=1.2​

Var(Y)​=1.2−(0.8)2=0.56​

Var(X+Y)​=0.49+0.56=1.05​

Standard deviation​=1.05​​

Solution:

(spoiler)

The standard deviation of total complaints is approximately 1.02.

Key points

Geometric distribution basics

  • Models number of trials until first success
  • Probability of success per trial: p
  • Probability formula: P(X=k)=(1−p)k−1⋅p

Formulas

  • Expected value: E(X)=p1​
  • Variance: Var(X)=p21−p​
  • Standard deviation: σ=p21−p​​

BITS conditions

  • Binary outcomes: success or failure
  • Independent trials
  • Trials not fixed; count until first success
  • Same probability p for each trial

Technology functions

  • geometpdf(p,x): probability first success on xth trial
  • geometcdf(p,x): probability first success on or before xth trial

Example applications

  • Define X as number of trials until first success (e.g., first third party voter)
  • Use E(X) for expected number of trials
  • Use σ for typical deviation from mean
  • Use P(X=k) or geometpdf for exact probabilities
  • Use geometcdf for cumulative probabilities
  • Use complement rule for probabilities of “at least” scenarios

Summary

  • Key formulas: P(X=k), E(X), Var(X), σ
  • Use technology for quick probability calculations
  • Always check BITS before applying geometric distribution

More from Probability and random variables

  • Law of large numbers
  • Introduction to probability
  • Expected value, variance, and standard deviation
  • Binomial distribution
  • Cumulative probability distribution